Influence of herringbone grooves on the performance of water-lubricated journal bearings: A novel AI-driven prediction and optimization approach

This study presents a novel framework for the design, analysis, artificial intelligence (AI)-based performance prediction, and numerical optimization of water-lubricated herringbone grooved journal bearings (HGJB), specifically tailored for underwater vehicle rotors and machinery operating at speeds up to 3000 rpm under radial loading conditions. Unlike traditional approaches, this research integrates advanced artificial intelligence tools-Artificial Neural Networks (ANN) and Adaptive Neuro-Fuzzy Inference Systems (ANFIS)-to optimize groove parameters, such as groove angle (Ga), groove depth (Gd), and groove number (Ng), while addressing multi-objective performance criteria. A numerical model was developed by solving the non-linear incompressible Reynolds equation using the central finite difference method (CFDM) to evaluate pressure distribution, load carrying capacity, frictional force, frictional coefficient, power loss, leakage rate, attitude angle, and frictional torque. This model generated data points for various operating and geometric parameters, forming a robust dataset for ANN training and enabling accurate performance prediction. ANFIS was employed to derive optimal values for groove parameters and operating conditions, demonstrating significant improvements in static performance metrics compared to conventional plain journal bearing. The novelty of this study lies in the systematic integration of AI tools with numerical modeling to enhance prediction accuracy and streamline multi-objective optimization, offering a comprehensive approach to HGJB design. This work not only provides new insights into the influence of groove parameters on hydrodynamic performance but also establishes a scalable framework for optimizing water-lubricated bearing systems in high-performance applications.